CROSS-REFERENCE TO RELATED APPLICATION
This application is a divisional application of and claims the priority benefit of U.S. application Ser. No. 15/878,836, filed on Jan. 24, 2018, now allowed. The entirety of the above-mentioned patent application is hereby incorporated by reference herein and made a part of this specification.
BACKGROUND OF THE INVENTION
Field of Invention
The present invention is related to a semiconductor device and a method of forming the same, and more generally to a FinFET device and a method of forming the same.
Description of Related Art
In the field of integrated circuit devices, the device dimensions are often reduced to attain a higher operating speed and a lower power consumption. However, with the ever-increasing level of integration of devices, the miniaturization of devices has almost reached its limit.
In recent years, a multi-gate structure such as a FinFET device is proposed to overcome the limitations imposed by the device miniaturization. Besides, strain engineering such as controlling the stress in the channel region of a transistor is also adopted. Source/drain regions are usually formed by growing epitaxial layers with an ion implantation process. However, different pattern densities in different regions across a wafer may affect the shapes and/or sizes of the grown epitaxial layers. When adjacent epitaxial layers are undesirably connected, a short circuit may occur, and the performance of the device may be decreased.
SUMMARY OF THE INVENTION
Accordingly, the present invention provides a FinFET device and a method of forming the same, in which epitaxial layers in different pattern density regions are formed separately with similar shapes and/or sizes, so the performance of the device is accordingly improved.
The present invention provides a method of forming a FinFET device that includes the following steps. A substrate is provided with a first region and a second region. First fins and second fins are formed on the substrate respectively in the first region and the second region. In an embodiment, a number of the first fins is different from a number of the second fins. A first dummy gate and a second dummy gate are formed respectively across the first fins and the second fins. A first spacer and a second spacer are sequentially formed on a sidewall of each of the first fins and the second fins. The first fins and the second fins are partially removed, so as to form first recesses in the first fins aside the first dummy gate and form second recesses in the second fins aside the second dummy gate. An oxidizing step is performed to the first spacers and the second spacers. Upper portions of the first spacers and the second spacers are removed. First epitaxial layers and second epitaxial layers are formed respectively in the first recesses and the second recesses. The first dummy gate and the second dummy gate are respectively replaced with a first gate and a second gate.
According to an embodiment of the present invention, the first spacers include SiOCN or SiCN, and the second spacers include SiN.
According to an embodiment of the present invention, the oxidizing step includes using O2, N2O, O3 or a combination thereof.
According to an embodiment of the present invention, removing the upper portions of the first spacers and the second spacers includes using a hydrogen-containing gas and a fluorine-containing gas.
According to an embodiment of the present invention, the hydrogen-containing gas includes NH3, N2H4, HN3 or a combination thereof, and the fluorine-containing gas includes NF3, SiF4, CF4, CH3F, CH2F2, CHF3, C2F6, C3F8 or a combination thereof.
According to an embodiment of the present invention, the number of the first fins is greater than the number of the second fins.
According to an embodiment of the present invention, the first region is a logic device region, and the second region is a SRAM device region.
According to an embodiment of the present invention, a maximum width of the first epitaxial layers is L1, a maximum width of the second epitaxial layers is L2, and (L2−L1)/L1 is equal to or less than about 1%.
According to an embodiment of the present invention, the first epitaxial layers and the second epitaxial layers include SiP or SiC.
According to an embodiment of the present invention, the method further includes, after removing the upper portions of the first spacers and the second spacers and before forming the first epitaxial layers and the second epitaxial layers, forming first buffer layers and second buffer layers respectively on surfaces of the first recesses and the second recesses.
According to an embodiment of the present invention, the first buffer layers and the second buffer layers include SiP or SiC, and a concentration of phosphorus or carbon of the first and second buffer layers is lower than a concentration of phosphorus or carbon of the first and second epitaxial layers.
The present invention further provides a FinFET device including a substrate, first fins and second fins, a first gate and a second gate, first epitaxial layers and second epitaxial layers. The substrate has a first region and a second region. The first fins and the second fins are disposed on the substrate respectively in the first region and the second region. In an embodiment, a number of the first fins is different from a number of the second fins. The first gate and the second gate are disposed on the substrate and respectively across the first fins and the second fins. The first epitaxial layers are disposed in first recesses of the first fins adjacent to the first gate. The second epitaxial layers are disposed in second recesses of the second fins adjacent to the second gate. In an embodiment, a maximum width of the first epitaxial layers is L1, a maximum width of the second epitaxial layers is L2, and (L2−L1)/L1 is equal to or less than about 1%.
According to an embodiment of the present invention, the number of the first fins is greater than the number of the second fins.
According to an embodiment of the present invention, the first region is a logic device region, and the second region is a SRAM device region.
According to an embodiment of the present invention, the first epitaxial layers and the second epitaxial layers include SiP or SiC.
According to an embodiment of the present invention, the FinFET device further includes first buffer layers on surfaces of the first recesses below the first epitaxial layers, and second buffer layers on surfaces of the second recesses below the second epitaxial layers.
According to an embodiment of the present invention, the first buffer layers and the second buffer layers include SiP or SiC, and a concentration of phosphorus or carbon of the first and second buffer layers is lower than a concentration of phosphorus or carbon of the first and second epitaxial layers.
According to an embodiment of the present invention, the FinFET device further includes a first spacer and a second spacer sequentially disposed on a sidewall of each of the first fins and second fins, and the first and second epitaxial layers laterally extend over the corresponding first and second spacers.
According to an embodiment of the present invention, tops of the first and second spacers in the first and second regions are substantially coplanar.
According to an embodiment of the present invention, the first spacers include SiOCN or SiCN, and the second spacers include SiN.
In view of the above, in the present invention, two steps of an oxide treatment and an oxide removal are performed prior to the formation of epitaxial layers in different regions. Such two steps remove the rough top portions of the spacers adjacent to the recesses, so the epitaxial layers grown from the recesses are formed with desired shapes and sizes, without being affected by the rough top portions of the spacers. Therefore, the performance of the device is accordingly improved.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, a preferred embodiment accompanied with figures is described in detail below.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.
FIG. 1 is a simplified top view of a FinFET device according to an embodiment of this invention.
FIG. 2A to FIG. 2G are schematic cross-sectional views of a method of forming a FinFET device in accordance with an embodiment of this invention.
DESCRIPTION OF EMBODIMENTS
Reference will now be made in detail to the present preferred embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the description to refer to the same or like parts.
FIG. 1 is a simplified top view of a FinFET device according to an embodiment of this invention, in which few elements such as fins and gates are shown for simplicity and clarity of illustration. FIG. 2A to FIG. 2G are schematic cross-sectional views of a method of forming a FinFET device in accordance with an embodiment of this invention, some of which are taken along the lines I-I′, I-II′ and III-III′ of FIG. 1.
Referring to FIG. 1 and FIG. 2A, a substrate 100 is provided. The substrate 100 can be a semiconductor substrate, such as a silicon-containing substrate. The substrate 100 has a first region 10 a and a second region 10 b. In an embodiment, the first region 10 a and the second region 10 b have different pattern densities. For example, a pitch, a dimension or a number of a feature in the first region 10 a is different from a pitch, a dimension or a number of a similar feature in the second region 10 b. In an embodiment, the first region 10 a and the second region 10 b are regions for P-type devices with different functions. In an embodiment, the first region 10 a is a logic device region, and the second region 10 b is a SRAM device region.
Thereafter, the substrate 100 is patterned to form first fins 102 a and second fins 102 b respectively in the first region 10 a and the second region 10 b. The method of forming the first fins 102 a and the second fins 102 b may include performing photolithography and etching processes. In an embodiment, the first fins 102 a and the second fins 102 b extend in a first direction D1, as shown in FIG. 1. In an embodiment, the number of the first fins 102 a is different from the number of the second fins 102 b. For example, the number of the first fins 102 a is greater than the number of the second fins 102 b. In an embodiment, the pitch of the first fins 102 a are different form the pitch of the second fins 102 b. For example, the pitch of the first fins 102 a are less than the pitch of the second fins 102 b. Specifically, the first region 10 a may be regarded as a dense region with more fins, and the second region 10 b may be regarded as a sparse region with less fins.
Afterwards, an isolation layer 103 is formed on the substrate 100. In an embodiment, the isolation layer 103 covers the lower portions of the first and second fins 102 a and 102 b while exposes the upper portions of the first and second fins 102 a and 102 b. The isolation layer 103 may include silicon oxide, and may be formed by depositing an isolation material layer over the substrate 100 and followed by an etching back process.
Still referring to FIG. 1 and FIG. 2A, a first dummy gate 104 a and a second dummy gate 104 b are formed over the substrate 100 and respectively across the first fins 102 a and the second fins 102 b. The first dummy gate 104 a and the second dummy gate 104 b may include a silicon-containing material (e.g., polysilicon) and the forming method thereof includes performing deposition processes and the subsequent photolithography and etching processes. In an embodiment, the first dummy gate 104 a and the second dummy gate 104 b extend in a second direction D2 different from (e.g., perpendicular to) the first direction D1, as shown in FIG. 1. In an embodiment, an oxidation process is performed to the first and second fins 102 a and 102 b before the formation of the first dummy gate 104 a and the second dummy gate 104 b, and thus, an interfacial layer (not shown) is formed on the surface of each of the first and second fins 102 a and 102 b. In an embodiment, a cap layer (not shown) is formed on each of the first dummy gate 104 a and the second dummy gate 104 b upon the process requirements.
Referring to FIG. 2B, in the first region 10 a, a first spacer 106 a and a second spacer 108 a are sequentially formed on the sidewall of first dummy gate 104 a and on the sidewalls of the first fins 102 a aside the first dummy gate 104 a. Similarly, in the second region 10 b, a first spacer 106 b and a second spacer 108 b are sequentially formed on the sidewall of second dummy gate 104 b and on the sidewalls of the second fins 102 b aside the second dummy gate 104 b. In an embodiment, the first spacers 106 a and 106 b include a material different form that of the second spacers 108 a and 108 b. For example, the first spacers 106 a and 106 b include SiOCN or SiCN, and the second spacers 108 a and 108 b include SiN. The method of forming the first spacers 106 a and 106 b and the second spacers 108 a and 108 b includes depositing first and second spacer material layers (not shown) on the substrate 100 and performing at least one anisotropic etching process to the first and second spacer material layers. In an embodiment, due to the pattern loading effect of the anisotropic etching process, the first spacers 106 a and 106 b and the second spacers 108 a and 108 b are formed with rough and irregular top surfaces, as shown in the enlarged local views of FIG. 2B. In an embodiment, the tops of the first and second spacers 106 a and 108 a in the first region 10 a (e.g., pattern dense region) is higher than the tops of the first and second spacers 106 b and 108 b in the second region 10 b (e.g., pattern sparse region).
Referring to FIG. 2C, in the first region 10 a, the first fins 102 a are partially removed to form first recesses 110 a in the first fins 102 a aside the first dummy gate 104 a. Specifically, the surface of one remaining first fin 102 a and the adjacent first spacer 102 a define a first recess 110 a. Similarly, in the second region 10 b, the second fins 102 b are partially removed to form second recesses 110 b in the second fins 102 b aside the second dummy gate 104 b. Specifically, the surface of one remaining second fin 102 b and the adjacent first spacer 102 b define a second recess 110 b.
In an embodiment, the first fins 102 a and the second fins 102 b are vertically etched to a predetermined depth and may be laterally etched to form a desired profile. Each of the vertical etching step and the lateral etching step includes a dry etching, a wet etching or both. In an embodiment, the first and second recesses 110 a and 110 b are formed with a bucket-like shape, a diamond-like shape, a middle-wide shape or the like.
In an embodiment, the surfaces of the remaining first and second fins 102 a and 102 b are higher than the surface of the isolation layer 103, but the present invention is not limited thereto. In another embodiment, the surfaces of the remaining first and second fins 102 a and 102 b may be as high as or lower than the surface of the isolation layer 103 upon the process requirements.
Referring to FIG. 2D, an oxidizing step 111 is preformed to the first spacers 106 a and 106 b and the second spacers 108 a and 108 b. In an embodiment, the oxidizing step 111 includes using O2, N2O, O3 or a combination thereof. The oxidizing step 111 is configured to partially oxidize the spacers. In an embodiment, when the first spacers 106 a and 106 b include SiOCN, and the second spacers 108 a and 108 b include SiN, merely the second spacers 108 a and 108 b are partially oxidized. In another embodiment, when the first spacers 106 a and 106 b include SiCN, and the second spacers 108 a and 108 b include SiN, both the first spacers 106 a and 106 b and the second spacers 108 a and 108 b are partially oxidized. In the oxidizing step 111, the oxidation time and/or the oxidation temperature can be controlled to determine the oxidized amounts of the spacers.
Referring to FIG. 2E, upper portions of the first spacers 106 a and 106 b and the second spacers 108 a and 108 b are removed. In an embodiment, the upper portions of the spacers indicate the oxidized portions of the spacers. In another embodiment, the upper portions of the spacers indicate the mixture of oxidized and non-oxidized portions of the spacers. In an embodiment, the removing step includes using a hydrogen-containing gas and a fluorine-containing gas. In an embodiment, the hydrogen-containing gas includes NH3, N2H4, HN3 or a combination thereof, and the fluorine-containing gas includes NF3, SiF4, CF4, CH3F, CH2F2, CHF3, C2F6, C3F8 or a combination thereof. In an embodiment, the removing step includes performing a dry etching process, such as a chemical oxide removal (COR), a SiCoNi™ etching or a combination thereof.
Upon the removing step, the rough and irregular upper portions of the first spacers 106 a and 106 b and the second spacers 108 a and 108 b are completely removed. In an embodiment, the remaining first spacers 107 a and 107 b and the remaining second spacers 109 a and 109 b are provided with substantially planar top surfaces. In an embodiment, the tops of the first and second spacers 107 a and 109 a in the first region 10 a (e.g., pattern dense region) is substantially coplanar with the tops of the first and second spacers 107 b and 109 b in the second region 10 b (e.g., pattern sparse region).
Referring to FIG. 2F, first epitaxial layers 114 a and second epitaxial layers 114 b are respectively formed in the first recesses 110 a and the second recesses 110 b. In an embodiment, the first epitaxial layers 114 a and the second epitaxial layers 114 b are formed by an epitaxial growth process with an ion implantation process. In an embodiment, in the first region 10 a, the lower portions of the first epitaxial layers 114 a are formed in the first recesses 110 a, and the upper portions of the first epitaxial layers 114 a are formed outside of the first recesses 110 a and laterally extend over the first and second spacers 107 a and 109 a. Similarly, in the second region 10 b, the lower portions of the second epitaxial layers 114 b are formed in the second recesses 110 b, and the upper portions of the second epitaxial layers 114 b are formed outside of the second recesses 110 b and laterally extend over the first and second spacers 107 b and 109 b. In an embodiment, the first and second recesses 110 a and 110 b are formed with a mushroom-like shape, a spade-like shape or the like.
In an embodiment, the first epitaxial layers 114 a and the second epitaxial layers 114 b include SiP or SiC, so as to provide tensile stresses to the channel regions. In an embodiment, the concentration of phosphorus or carbon of each of the first epitaxial layers 114 a and the second epitaxial layers 114 b is substantially constant. In another embodiment, the concentration of phosphorus or carbon of each of the first epitaxial layers 114 a and the second epitaxial layers 114 b is gradually varying from inside to outside and/or from bottom to top.
In an embodiment, in the first region 10 a, each of the first epitaxial layers 114 a are formed separately. Similarly, in the second region 10 b, each of the second epitaxial layers 114 b are formed separately.
It is noted that, due to the pattern loading effect of the epitaxial growth process, the sizes of epitaxial layers in different pattern density regions are usually significantly different from each other, and thus, the size difference between the epitaxial layers in different pattern density regions is large (e.g., about 20% or more). However, such issue is not observed in the present invention. Specifically, the epitaxial layers are grown after the rough top portions of the adjacent spacers are removed, so the grown rates of the epitaxial layers are not affected by the rough top portions of the adjacent spacers, and thus, the size difference between the epitaxial layers in different pattern density regions is small (e.g., about 1% or less).
In an embodiment, the first epitaxial layers 114 a and the second epitaxial layers 114 b are formed with similar sizes. In an embodiment, the size of the second epitaxial layers 114 b is slightly greater than the size of the first epitaxial layers 114 a. The size may be width, height or both. For example, the maximum width of the second epitaxial layers 114 b is slightly greater than the maximum width of the first epitaxial layers 114 a by about 1% or less. In an embodiment, each of the first epitaxial layers 114 a and the second epitaxial layers 114 b has a maximum with of about 410 angstroms to 480 angstroms, and the size difference between the first epitaxial layers 114 a and the second epitaxial layers 114 b is about 4 angstroms or less.
In an embodiment, after removing the upper portions of the first and second spacers (as shown in FIG. 2F) and before forming the first and second epitaxial layers (as shown in FIG. 2G), first buffer layers 112 a and second buffer layers 112 b are respectively formed on the surfaces of the first recesses 110 a and the second recesses 110 b. In an embodiment, the first buffer layers 112 a and the second buffer layers 112 b include SiP or SiC, and the concentration of phosphorus or carbon of the first and second buffer layers 112 a and 112 b is lower than the concentration of phosphorus or carbon of the first and second epitaxial layers 114 a and 114 b.
Referring to FIG. 2G, the first dummy gate 104 a and the second dummy gate 104 b are respectively replaced with a first gate 120 a and a second gate 120 b. In an embodiment, the first gate 120 a and the second gate 120 b extend in a second direction D2 different from (e.g., perpendicular to) the first direction D1, as shown in FIG. 1. In an embodiment, a dielectric layer 115 is formed over the substrate 100, surrounds the sidewalls of the first and second dummy gates 104 a and 104 b and exposes the tops of the first and second dummy gates 104 a and 104 b. The dielectric layer 115 may include silicon oxide, a low-k material, a suitable insulating material or a combination thereof. Thereafter, the first and second dummy gates 104 a and 104 b are removed to form first and second gate trenches in the dielectric layer 115.
Afterwards, first and second gates 120 a and 120 b are respectively formed in the first and second gate trenches. In an embodiment, in the first region 10 a, the first gate 120 a includes a first gate dielectric layer 116 a and a first work function metal layer 118 a. Similarly, in the second region 10 b, the second gate 120 b includes a second gate dielectric layer 116 b and a second work function metal layer 118 b. Each of the first and second gate dielectric layers 116 a and 116 b includes an interfacial layer, a high-k layer or a combination thereof. The interfacial layer includes silicon oxide, silicon oxynitride or the like. The high-k layer can be metal oxide including hafnium oxide (HfO2), hafnium silicon oxide (HfSiO4), hafnium silicon oxynitride (HfSiON), aluminum oxide (Al2O3), lanthanum oxide (La2O3), tantalum oxide (Ta2O5), yttrium oxide (Y2O3), zirconium oxide (ZrO2), strontium titanate oxide (SrTiO3), zirconium silicon oxide (ZrSiO4), hafnium zirconium oxide (HfZrO4), strontium bismuth tantalate, (SrBi2Ta2O9, SBT), lead zirconate titanate (PbZrxTi1-xO3, PZT, x>0), barium strontium titanate (BaxSr1-xTiO3, BST, x>0) or a combination thereof. In an embodiment, for a P-type device, each of the first and second work function layers 118 a and 118 b includes titanium nitride (TiN), titanium carbide (TiC), tantalum nitride (TaN), tantalum carbide (TaC), tungsten carbide (WC) or aluminum titanium nitride (TiAlN) or a combination thereof. The FinFET device 10 of the invention is thus completed.
The structure of the FinFET device of the invention is descried below with reference to FIG. 2G. As shown in FIG. 2G, the FinFET device 10 includes a substrate 100, first fins 102 a and second fins 102 b, a first gate 120 a and a second gate 120 b, first epitaxial layers 114 a and second epitaxial layers 114 b. The substrate 100 has a first region 10 a and a second region 10 b. In an embodiment, the first region 10 a is a logic device region, and the second region 10 b is a SRAM device region. The first fins 102 a and the second fins 102 b are disposed on the substrate 100 respectively in the first region 10 a and the second region 10 b. In an embodiment, the number of the first fins 102 a is different from (e.g., greater than) the number of the second fins 102 b. The first gate 120 a and the second gate 120 b are disposed on the substrate 100 and respectively across the first fins 102 a and the second fins 102 b. The first epitaxial layers 114 a are disposed in first recesses 110 a of the first fins 102 a adjacent to the first gate 120 a. The second epitaxial layers 114 b are disposed in second recesses 110 b of the second fins 102 b adjacent to the second gate 120 b. In an embodiment, the first epitaxial layers 114 a and the second epitaxial layers 114 b include SiP or SiC. In an embodiment, the maximum width of the first epitaxial layers 114 a is L1, the maximum width of the second epitaxial layers 114 b is L2, and (L2−L1)/L1 is equal to or less than about 1%. For example, the ratio of (L2−L1)/L1 can be, for example but not limited to, about 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.010, including any range between any two of the preceding values.
In an embodiment, each of the first and second epitaxial layers 114 a and 114 b has a mushroom-like shape. Specifically, each of the first and second epitaxial layers 114 a and 114 b has a lower portion (or mushroom stem portion) and an upper portion (or mushroom cap portion) over the lower portion. In an embodiment, the upper portions of the first and second epitaxial layers 114 a and 114 b has a diamond-like shape or a middle-wide profile.
In an embodiment, the FinFET device 10 further includes first spacers 107 a and 107 b and second spacers 109 a and 109 b surround the lower portions (or mushroom stem portions) and below the upper portions (or mushroom cap portions) of the first and second epitaxial layers 114 a and 114 b. Specifically, in the first region 10 a, a first spacer 107 a and a second spacer 109 a are sequentially disposed on the sidewall of each of the first fins 102 a, and the first epitaxial layers 114 a laterally extend over the corresponding first and second spacers 107 a and 109 a. Similarly, in the second region 10 b, a first spacer 107 ab and a second spacer 109 b are sequentially disposed on the sidewall of each of the second fins 102 b, and the second epitaxial layers 114 b laterally extend over the corresponding first and second spacers 107 b and 109 b. In an embodiment, the tops of the first spacers 107 a and 107 b and the second spacers 109 a and 109 b in the first and second regions 10 a and 10 b are substantially coplanar. In an embodiment, the first spacers 107 a and 107 b include SiOCN or SiCN, and the second spacers 109 a and 109 b include SiN.
In an embodiment, the FinFET device 10 further includes first buffer layers 112 a and second buffer layers 112 b. The first buffer layers 112 a are disposed on surfaces of the first recesses 110 a below the first epitaxial layers 114 a. The second buffer layers 112 b are disposed on surfaces of the second recesses 110 b below the second epitaxial layers 114 b. In an embodiment, the first buffer layers 112 a and the second buffer layers 112 b include SiP or SiC, and the concentration of phosphorus or carbon of the first and second buffer layers 112 a and 112 b is lower than the concentration of phosphorus or carbon of the first and second epitaxial layers 114 a and 114 b.
In view of the above, in the present invention, before epitaxial layers are grown from recesses in different pattern density regions, the rough top portions of the spacers adjacent to the recesses are removed with two steps of an oxidizing treatment and an oxide removal. Since the rough top portions of the spacers are removed without affecting the grown rates of the epitaxial layers, the epitaxial layers in different pattern density regions are formed with desired shapes and sizes. Therefore, the performance of the device is accordingly improved.
The present invention has been disclosed above in the preferred embodiments, but is not limited to those. It is known to persons skilled in the art that some modifications and innovations may be made without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be defined by the following claims.